AI-CABB: An AI-Enabled Integrated Pest Detection System for Enhancing Cabbage Cultivation Using Machine Learning

One of the most important vegetable crops on the planet, cabbage, faces a threat from destructive insects that reduce yields and harm businesses. As a commonly grown crop, cabbage is vulnerable to pests including aphids, whiteflies, and diamondback moths. Sustainable solutions are necessary, as conventional pest management techniques like the use of pesticides endanger both the environment and human health. Conventional control techniques, such as physical inspection and chemical treatments, prove to be time-consuming, labor-intensive, and hazardous to the environment and human health. The transformative potential of ML lies in its ability to accurately identify and manage pests while mitigating environmental harm. The emphasis on machine learning-based identification of hazardous insects attacking cabbage fills an existing gap. This work aims to address the detection gap for damaging insects attacking cabbage using machine learning (ML), specifically in the diagnosis and management of crop concerns. Additionally, one of the key accomplishments of this work is the creation of a dataset made up of four different insect classes. Later, the detection model, YOLOv5, is trained and validated using this dataset. The resulting machine learning model enables the identification of harmful insects in real-time, potentially revolutionizing the control of insect pests in agriculture. Given the importance of agriculture to economies, this research acknowledges the challenges farmers face, including pest infestations that diminish crop productivity. The outcomes of real-time identification of different harmful insects support improved pest management techniques and provide insights for future research.